Yield Estimation of Paddy Rice Based on Satellite Imagery: Comparison of Global and Local Regression Models

Yi-Shiang Shiu,Yung-Chung Chuang

doi:10.3390/rs11020111

Abstract

Precisely estimating the yield of paddy rice is crucial for national food security and development evaluation. Rice yield estimation based on satellite imagery is usually performed with global regression models; however, estimation errors may occur because the spatial variation is not considered. Therefore, this study proposed an approach estimating paddy rice yield based on global and local regression models. In our study area, the overall per-field data might not available because it took lots of time and manpower as well as resources. Therefore, we gathered and accumulated 26 to 63 ground survey sample fields, accounting for about 0.05% of the total cultivated areas, as the training samples for our regression models. To demonstrate whether the spatial autocorrelation or spatial heterogeneity exists and dominates the estimation, global models including the ordinary least squares (OLS), support vector regression (SVR), and the local model geographically weighted regression (GWR) were used to build the yield estimation models. We obtained the representative independent variables, including 4 original bands, 11 vegetation indices, and 32 texture indices, from SPOT-7 multispectral satellite imagery. To determine the optimal variable combination, feature selection based on the Pearson correlation was used for all of the regression models. The case study in Central Taiwan rendered that the error rate was between 0.06% and 13.22%. Through feature selection, the GWR model’s performance was more relatively stable than the OLS model and nonlinear SVR model for yield estimation. Where the GWR model considers the spatial autocorrelation and spatial heterogeneity of the relationships between the yield and the independent variables, the OLS and nonlinear SVR models lack this feature; this led to the rice yield estimation of GWR in this study be more stable than those of the other two models.

Highlights

With the improvement of agricultural science and technology, pests, bacteria, and environmental factors are having less of an effect on the growth of rice
Among the 2016 first-cultivation yield estimations of the three yield estimation models, the results of the rice yield were most unfavorable for the Combination 1 model, and the geographically weighted regression (GWR) model could not even be calculated
In Combinations 2 and 3, the support vector regression (SVR) rice yield errors were no different, and the ordinary least squares (OLS) model error was reduced through the process of feature selection

Summary

Introduction

With the improvement of agricultural science and technology, pests, bacteria, and environmental factors are having less of an effect on the growth of rice. On the authority of the FAOSTAT database from Food and Agriculture Organization of the United Naitons, the harvested area and yield of paddy rice reached a historic high of 145.4 million hectares and 672.5 million tons in Asia in 2013. The aforementioned data infers the growth of the rice yield per unit area, and reflects the development of the national economy. Precisely mapping the cultivation area and estimating the yield of paddy rice are both crucial for national food security and national development evaluation. Concerning the paddy rice yield, ground-based field survey is still an essential process for regional and national level estimation, this necessary survey is usually thought of as being time-consuming, subjective [2], and costly [3]. In many studies, satellite data selection, application, analysis, and verification methods differ from one another; effectively attenuating the limitations of previous studies is an essential aspect of estimation

Methods

Results

Discussion

Conclusion